Self-adaptive prediction correction method and system for position of motor rotor

Abstract

The invention discloses a self-adaptive prediction correction method and system for the position of a motor rotor, and belongs to the technical field of motor driving control. The method comprises the following steps: firstly, acquiring rotor positions of a motor in current and historical multiple control periods; then, a first prediction position is calculated in parallel based on the motor constant speed state according to the latest two period positions, and a second prediction position is calculated based on the motor constant acceleration state according to the latest three period positions; according to the prediction errors of the two states in the previous period, the adaptive weight coefficient of the current period is dynamically calculated; and finally, the corrected rotor position is obtained through weighted fusion. Through a double-model adaptive fusion mechanism, high-precision position prediction is realized in a full working condition range, position detection lag caused by control delay is effectively overcome, a d-q axis decoupling effect is remarkably improved, current oscillation and torque ripple are inhibited, and dynamic performance and operation stability of a motor control system are comprehensively improved.

Description

Technical Field

[0001] This invention relates to the field of motor drive control technology, specifically to an adaptive prediction and correction method and system for motor rotor position. Background Technology

[0002] Permanent magnet synchronous motors are widely used in high-performance drive applications due to their high efficiency and high power density. Field-oriented control is the mainstream technology for achieving high-performance control of these motors. This technology decouples AC quantities into DC components (dq) through coordinate transformation, thereby enabling independent control of motor torque and flux linkage.

[0003] The accuracy of FOC control heavily relies on the precise detection of the rotor's real-time position. When using a digital processor such as a microcontroller (MCU) to execute the FOC algorithm, the control flow is typically completed within periodic interrupts. However, there is an unavoidable processing time delay from the moment the interrupt triggers and acquires the rotor position to the time the controller completes a series of calculations (such as PI regulation and SVPWM generation) and finally applies the voltage command to the inverter. This delay is particularly pronounced when the motor is operating at high speeds or has a large number of pole pairs. For example, a 14-pole motor at 3000 RPM has an extremely high electrical angular velocity. If the controller interrupt frequency is 10kHz, a delay of just one control cycle can lead to an electrical angular deviation of up to 20-30 degrees. This means that the control command calculated by the controller based on an "outdated" rotor position is applied to a new position where the rotor has already moved.

[0004] The lag in position detection using a controller can directly lead to the following serious problems: First, control inaccuracies and efficiency degradation occur: the actual voltage vector deviates from the ideal q-axis, making it impossible to achieve maximum torque-to-current ratio control (e.g., id=0 control), thus reducing the motor's energy conversion efficiency. Second, system instability arises: position errors disrupt the decoupling premise of the dq-axis, introducing unnecessary coupling terms, which in turn cause torque pulsation and current oscillations, impairing the smoothness of motor operation. Existing compensation methods, such as uniform linear extrapolation based on the previous moment's speed, are simple but only effective when the motor is running at a constant speed. Once the motor enters dynamic conditions such as acceleration or deceleration, the prediction error of this method increases significantly, failing to meet the dynamic response and accuracy requirements of high-performance servo systems.

[0005] Therefore, how to effectively compensate for the position error caused by control delay, especially by providing an accurate and adaptive prediction method across the entire operating range of the motor (including variable speed mode), is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] To address the problem in existing technologies where rotor position detection lags due to control system processing delays, thereby affecting motor control performance, this invention provides an adaptive prediction and correction method and system for motor rotor position. By accurately compensating for this delay, the actual rotor position is tracked in real time.

[0007] This invention is achieved through the following technical solution: In a first aspect, this application provides an adaptive prediction and correction method for the rotor position of a motor, comprising the following steps: Obtain the rotor position of the motor in the current control cycle and at least the previous two historical control cycles; Based on the constant speed state of the motor, the first predicted position is calculated according to the rotor position of the most recent two control cycles; Based on the constant acceleration state of the motor, the second predicted position is calculated according to the rotor position of the most recent three control cycles; Based on the prediction error of the motor constant speed state and the motor constant acceleration state in the previous control cycle, the adaptive weighting coefficient of the current cycle is dynamically calculated. The first predicted position and the second predicted position are weighted and fused using the adaptive weighting coefficients to obtain the corrected final rotor position.

[0008] Preferably, the calculation of the first predicted position based on the constant speed state of the motor specifically includes: Calculate the average angular velocity based on the rotor position difference between the k-th control cycle and the (k-1)-th control cycle; The first predicted position is obtained by extrapolating the system delay time based on the average angular velocity; Preferably, the calculation of the second predicted position based on the constant acceleration state of the motor specifically includes: According to the k, k-1, k-2 The rotor position during the period is estimated, taking into account the instantaneous angular velocity caused by acceleration. The second predicted position is obtained by extrapolating the system delay time forward based on the instantaneous angular velocity.

[0009] Preferably, the dynamic calculation of the adaptive weighting coefficients specifically includes: According to the k Period and the k-1 The rotor position in each cycle determines the actual reference position for the (k-1)th cycle; Calculate the first absolute error between the first predicted position and the actual reference position; Calculate the second absolute error between the second predicted position and the actual reference position; The adaptive weighting coefficients are calculated based on the first absolute error and the second absolute error.

[0010] Preferably, the firstk-1 The method for determining the actual reference position of the control cycle is as follows: For the first k Control cycle and the first k-1 The actual rotor position during the control cycle is linearly interpolated to obtain the actual reference position.

[0011] Preferably, the adaptive weighting coefficients are calculated using the following method: The error ratio is calculated based on the absolute error between the constant speed state and the constant acceleration state of the motor in the previous control cycle. The error ratio is input into an S-shaped function for normalization to generate the adaptive weight coefficient for the current period.

[0012] Preferably, the weighted fusion to obtain the final rotor position specifically involves:

[0013] in, The first predicted position, The second predicted position, For adaptive weighting coefficients, ...

[0014] Preferably, the method further includes determining the final rotor angular velocity. The method is as follows:

[0015] in, The average angular velocity, It is the instantaneous angular velocity. These are adaptive weighting coefficients.

[0016] Secondly, this application provides an adaptive prediction and correction system for motor rotor position, comprising: The acquisition module is used to obtain the rotor position of the motor in the current control cycle and at least the previous two historical control cycles; The prediction module is used to calculate the first predicted position based on the rotor position of the most recent two control cycles, based on the constant speed state of the motor. Based on the constant acceleration state of the motor, the second predicted position is calculated according to the rotor position of the most recent three control cycles; The weighting module is used to dynamically calculate the adaptive weighting coefficient for the current cycle based on the prediction error of the motor constant speed state and the motor constant acceleration state in the previous control cycle. The correction module is used to perform weighted fusion of the first predicted position and the second predicted position using the adaptive weighting coefficients to obtain the corrected final rotor position.

[0017] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the adaptive prediction and correction method for the motor rotor position as described above.

[0018] Compared with the prior art, the present invention has the following beneficial technical effects: The adaptive prediction and correction method for motor rotor position provided by this invention first acquires historical rotor position data from multiple consecutive control cycles, and independently predicts the delayed rotor position based on constant speed and constant acceleration assumptions, respectively, forming complementary prediction results. Its key improvement lies in the introduction of a dynamic weight adjustment strategy. By retrospectively comparing the prediction error performance of the two state models in the previous control cycle, the optimal fusion weight for the current cycle is automatically calculated, thereby achieving intelligent correction of the prediction results. The innovation of this strategy lies in its breakthrough overcoming the technical bottleneck of insufficient adaptability of traditional single prediction models under variable speed conditions, enabling rotor position prediction to maintain extremely high estimation accuracy and robustness throughout the entire operating range of the motor—including various operating states such as constant speed, acceleration, and deceleration. Through this dual-model adaptive fusion strategy, this invention effectively overcomes the position detection lag problem caused by processing delays in digital control systems, significantly improves the decoupling effect of the dq axis in field-oriented control, and thus effectively suppresses current oscillations and torque pulsations. This technical solution not only improves the real-time performance and accuracy of position detection, but also optimizes the dynamic response performance and operational stability of the motor control system as a whole, providing an effective solution for high-performance motor drive applications.

[0019] This application also proposes an adaptive prediction and correction system for motor rotor position, an electronic device, and a computer storage medium, which possess all the advantages of the aforementioned adaptive prediction and correction method for motor rotor position. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the adaptive prediction and correction method for motor rotor position according to the present invention; Figure 2 This is a block diagram of the motor control system of the present invention; Figure 3 This is a schematic diagram of rotor position prediction according to an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0024] An adaptive prediction and correction method for motor rotor position includes the following steps: Obtain the rotor position of the motor in the current control cycle and at least the previous two historical control cycles; Based on the constant speed state of the motor, the first predicted position is calculated according to the rotor position of the most recent two control cycles; Based on the constant acceleration state of the motor, the second predicted position is calculated according to the rotor position of the most recent three control cycles; Based on the prediction error of the motor constant speed state and the motor constant acceleration state in the previous control cycle, the adaptive weighting coefficient of the current cycle is dynamically calculated. The first predicted position and the second predicted position are weighted and fused using the adaptive weighting coefficients to obtain the corrected final rotor position.

[0025] This method first obtains the rotor position of the motor at the start of the current and multiple historical control cycles using a position sensor. Based on this position sequence, the method calculates two predicted rotor positions in parallel, taking into account system delays. First, the first predicted position is based on the constant speed state of the motor, which assumes that the motor is running at a constant speed recently. Its calculation depends on the position information of the last two control cycles. Secondly, there is a second predicted position based on the constant acceleration state of the motor. This state assumes that the motor is currently operating with uniform acceleration, and its calculation depends on the position information from the last three control cycles. These two states correspond to the uniform speed and variable speed operating conditions of the motor, respectively, and each exhibits an advantage in prediction accuracy under these conditions.

[0026] The key innovation of this invention lies in the introduction of an adaptive fusion mechanism, rather than using a fixed single state for prediction. This mechanism achieves intelligent fusion by evaluating the historical prediction performance of two states in the previous control cycle. Specifically, the system calculates the error between the predicted value of each of the two states and an actual reference position calibrated by posterior interpolation. By comparing the relative magnitudes of these two historical prediction errors, it can be determined which state in the previous cycle better matches the actual operating conditions of the motor.

[0027] Based on this judgment, the system dynamically calculates an adaptive weighting coefficient K suitable for the current control cycle. This coefficient is used to weight and fuse the two predicted positions calculated in parallel for the current cycle, thereby outputting an optimal, corrected final rotor position. This mechanism makes the method adaptive: when historical errors indicate that the motor is running at a constant speed, the weighting coefficient K will automatically tend to trust the prediction result of the constant speed state; conversely, when historical errors indicate that the motor is accelerating or decelerating, the weighting coefficient K will automatically tend to trust the prediction result of the constant acceleration state.

[0028] Example 1 See Figure 1 The figure depicts the rotor electrical angles. The variation over time t. The motor system operates with a fixed control cycle. exist k-2 , k-1,k The rotor position is sampled at discrete time intervals. k There is a processing delay between when an interrupt is triggered and when the controller completes the calculation and applies the voltage. t1 The objective of this invention is to... k Based on the historical position readings of the current time and previous times, accurate predictions can be made. t1 The actual position reached by the rotor after the delay time. .

[0029] The following is a detailed description of an adaptive prediction and correction method for motor rotor position according to this embodiment, including the following steps: Step 1: Obtain the rotor position of the motor in the current control cycle and in multiple historical control cycles; The rotor position is sampled in real time by position sensors such as encoders to obtain rotor position data for the current cycle and at least the previous two or three control cycles.

[0030] For example, obtaining the current control cycle k Control cycle k- 1. Control cycle k- Rotor position 2 , and .

[0031] Step 2: Based on the constant speed state of the motor, calculate the first predicted position according to the rotor position of the most recent two control cycles; Assuming the motor moves at a constant speed in the most recent cycle. Based on this assumption, the control cycle is... k With control cycle k- Calculate the average angular velocity based on the rotor position difference. The formula for its calculation is:

[0032] Based on average angular velocity Calculation in k After a moment t1 The first predicted position of the rotor at the specified time point .

[0033]

[0034] in, t1 To delay time, k The time is the control cycle k A moment of revelation.

[0035] Step 3: Based on the constant acceleration state of the motor, calculate the second predicted position according to the rotor position of the most recent three cycles; Assume the motor is undergoing uniform acceleration during the last two control cycles.

[0036] First, calculate the average speed of the rotor for the two most recent control cycles; Then, based on the assumption of uniform acceleration, the instantaneous angular velocity at the current moment k, taking into account the effect of acceleration, is estimated according to the average velocity of the two control cycles. ; Finally, based on the instantaneous angular velocity Calculation in k After a moment t1 The second predicted position of the rotor at the time point .

[0037]

[0038]

[0039] Step 4: Based on the prediction error of the constant speed state and constant acceleration state in the previous control cycle, dynamically calculate the adaptive weighting coefficient for the current cycle. First, according to the control cycle k and control cycle k- Rotor position 1 and Determine the control cycle k-1. Actual reference position ; Specifically, regarding rotor position and Perform linear interpolation to determine the control period. k- 1. Actual reference position ;

[0040] Secondly, calculate the actual reference position. Respectively with the first predicted position Second predicted position absolute error between and The relative magnitude of the two historical errors directly reflects the motor's operating status at that time.

[0041]

[0042]

[0043] Finally, adaptive weighting coefficients for the current period are dynamically generated based on the two absolute errors and the Sigmoid function. ,as follows:

[0044] This adaptive predictive correction method, after obtaining the predicted positions at two operating speeds, needs to fuse the two rotor predicted positions and evaluate the performance of the predicted positions in the previous control cycle. The adaptive weighting coefficients... It can adaptively adjust based on historical errors: when the motor tends to a constant speed ( (smaller) A value approaching 0 means the algorithm trusts the predictions derived from the constant-speed state more in the current cycle; conversely, when the motor is in a variable-speed state ( (smaller) If the value approaches 1, the prediction derived from the constant acceleration state is more reliable; when the two states behave similarly, It then approaches 0.5, at which point each takes half the weight.

[0045] Step 5: Using the adaptive weighting coefficients, the first predicted position and the second predicted position are weighted and fused to obtain the corrected final rotor position and the final rotor angular velocity. and rotor position The calculation method is as follows:

[0046]

[0047] This adaptive predictive correction method can intelligently determine the motor's operating state in each control cycle and assign higher weights to more reliable predicted positions, thereby outputting highly accurate rotor position and speed under all operating conditions. It is worth mentioning that, due to the speed... It is obtained through a smooth weighted average rather than by directly differentiating the position signal. Therefore, it can effectively reduce the speed signal glitches caused by differentiation operations in traditional methods, making the speed loop control more stable.

[0048] This adaptive prediction and correction method for motor rotor position aims to solve the position detection lag problem caused by control system delays during high-speed motor operation. The method establishes a constant speed prediction model and a constant acceleration prediction model in parallel to predict the rotor position after considering the delay time. Its core innovation lies in dynamically generating an adaptive weighting coefficient K by calculating and comparing the prediction errors of the two models in the previous control cycle. This weighting coefficient is used to weight and fuse the prediction results of the two models in the current cycle, thereby outputting the final corrected position. When the motor approaches a constant speed, the method automatically trusts the constant speed model; when the motor is in a variable speed state, it automatically trusts the constant acceleration model. This invention can achieve high-precision position prediction under all operating conditions, effectively suppressing torque pulsation and current oscillation, and significantly improving the performance and stability of the motor control system.

[0049] Example 2 A field-oriented control (FOC) system for a permanent magnet synchronous motor, wherein the system performs an adaptive predictive correction method for the rotor position of the motor during operation.

[0050] An adaptive predictive correction method for motor rotor position is applied to the field-oriented control (FOC) system of a permanent magnet synchronous motor. Please refer to [link / reference]. Figure 3 This is a typical FOC system block diagram, in which the core of this invention, the adaptive position and speed predictor, is integrated into the control loop. When this adaptive position and speed predictor operates, it executes an adaptive prediction and correction method for the motor rotor position. The input to this predictor is the original position directly read from the encoder. Its output is the real-time position after precise prediction and correction. and real-time rotation speed Corrected position The data is fed into the Park transform and inverse transform module to ensure accurate decoupling of the dq axis; the corrected rotational speed... This signal is then sent to the speed controller as a feedback signal.

[0051] Example 3 Correspondingly, this application also provides an adaptive prediction and correction system for motor rotor position, including: The acquisition module is used to obtain the rotor position of the motor in the current control cycle and at least the previous two historical control cycles; The prediction module is used to calculate the first predicted position based on the rotor position of the most recent two control cycles, based on the constant speed state of the motor. Based on the constant acceleration state of the motor, the second predicted position is calculated according to the rotor position of the most recent three control cycles; The weighting module is used to dynamically calculate the adaptive weighting coefficient for the current cycle based on the prediction error of the motor constant speed state and the motor constant acceleration state in the previous control cycle. The correction module is used to perform weighted fusion of the first predicted position and the second predicted position using the adaptive weighting coefficients to obtain the corrected final rotor position.

[0052] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another device, or some features may be ignored or not executed. The modules described as separate components may or may not be physically separated. The components shown as modules may be one or more physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0053] Furthermore, in the various embodiments of the present invention, the modules can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0054] An electronic device provided in this application includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the adaptive prediction and correction method for the motor rotor position as described in any of the above embodiments.

[0055] Another electronic device provided in this application embodiment may further include: an input port connected to a processor for transmitting multimodal data collected by an external acquisition device to the processor; a display unit connected to the processor for displaying the processor's processing results to the outside world; and a communication module connected to the processor for enabling communication between the electronic device and the outside world. The display unit may be a display panel, a laser scanning display, etc.; the communication method adopted by the communication module includes, but is not limited to, Mobile High Definition Link (HML), Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), and wireless connection (including Wi-Fi, Bluetooth, Bluetooth Low Energy, and IEEE 802.11s-based communication technology).

[0056] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the adaptive prediction and correction method for motor rotor position as described in any of the above embodiments.

[0057] For descriptions of relevant parts of the adaptive prediction and correction system, electronic device, and computer-readable storage medium for motor rotor position provided in this application, please refer to the detailed descriptions of the corresponding parts in the adaptive prediction and correction method for motor rotor position provided in this application, which will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0058] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method of adaptive prediction correction of motor rotor position, characterized in that, The method comprises the following steps: acquiring rotor positions of the motor in a current control period and at least two previous historical control periods; calculating a first predicted position based on a constant speed state of the motor according to rotor positions in the last two control periods; calculating a second predicted position based on a constant acceleration state of the motor according to rotor positions in the last three control periods; dynamically calculating an adaptive weight coefficient of the current period according to prediction errors of the motor constant speed state and the motor constant acceleration state in the last control period; weighting and fusing the first predicted position and the second predicted position using the adaptive weight coefficient to obtain a corrected final rotor position.

2. A method of adaptive prediction-correction of motor rotor position according to claim 1, characterized in that, The calculation of the first predicted position based on the constant speed state of the motor specifically comprises: calculating an average angular velocity according to a rotor position difference between the kth control period and the (k-1)th control period; extrapolating the system delay time according to the average angular velocity to obtain the first predicted position.

3. A method of adaptive prediction-correction of motor rotor position according to claim 1, characterized in that, The calculation of the second predicted position based on the constant acceleration state of the motor specifically comprises: According to the first k, k-1, k-2 the instantaneous angular velocity considering the acceleration effect is estimated according to the periodic rotor position; extrapolating the system delay time according to the instantaneous angular velocity to obtain the second predicted position.

4. The adaptive prediction-correction method of a motor rotor position according to claim 1, wherein, The dynamic calculation of the adaptive weight coefficient specifically comprises: According to the first k The rotor position of the kth cycle and the rotor position of the k-1th cycle are determined. k-1 The actual reference position of the k-1th cycle is determined according to the rotor position of the kth cycle and the rotor position of the k-1th calculating a first absolute error of the first predicted position and the actual reference position; calculating a second absolute error of the second predicted position and the actual reference position; calculating the adaptive weight coefficient according to the first absolute error and the second absolute error.

5. A method of adaptive prediction-correction of motor rotor position according to claim 4, characterized in that, The first k-1 The determination of the actual reference position of the control cycle is as follows: for the first k the control period and the actual position of the rotor for the first k-1 linearly interpolating the actual position of the rotor for the control period to obtain the actual reference position.

6. A method of adaptive prediction-correction of motor rotor position according to claim 4, characterized in that, The calculation method of the adaptive weight coefficient is as follows: calculating an error ratio value based on absolute errors of the motor constant speed state and the motor constant acceleration state in the last control period; inputting the error ratio value into an S-type function for normalization processing to generate the adaptive weight coefficient of the current period.

7. The adaptive prediction-correction method of a motor rotor position according to claim 1, wherein, The weighting fusion to obtain the final rotor position specifically comprises: wherein, is a first predicted position, is a second predicted position, is an adaptive weight coefficient.

8. The adaptive prediction-correction method of a motor rotor position according to claim 1, wherein, The method also includes determining a final rotor angular velocity The method is as follows: wherein is the average angular velocity, is the instantaneous angular velocity, is the adaptive weight coefficient.

9. An adaptive prediction-correction system for motor rotor position, characterized by, comprises: a collection module for acquiring rotor positions of the motor in a current control period and at least two previous historical control periods; a prediction module for calculating a first predicted position based on a constant speed state of the motor according to rotor positions in the last two control periods; calculating a second predicted position based on a constant acceleration state of the motor according to rotor positions in the last three control periods; a weight module for dynamically calculating an adaptive weight coefficient of the current period according to prediction errors of the motor constant speed state and the motor constant acceleration state in the last control period; a correction module for weighting and fusing the first predicted position and the second predicted position using the adaptive weight coefficient to obtain a corrected final rotor position.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by a processor to implement the adaptive prediction correction method of the motor rotor position according to any one of claims 1 to 8.

Images